"... In PAGE 7: ... Although the polyprotein 1ab has been assigned only replicase activity, our analysis implies that the replicase activity is associated with Sars2628 (C terminal of ORF 1ab) fragment. The complete 1ab polyprotein contains 6 functional signatures of which polyprotein 1a contains signatures associated with metabolic enzymes ( Table5 a). Functions were assigned to the polyproteins on the basis of peptides (length 7 or more amino acids) occurring in proteins having similar functions in at least 5 different organisms.... In PAGE 7: ... Other predicted genes/protein coding regions contain peptides which occur in fewer genomes. Based on these peptides we suggest functions, albeit with lesser confidence ( Table5 b). The biological relevance ... In PAGE 15: ... Table5 (a): Functional assignment of polyproteins in SARS (Urbani) Genome using PLHOST S.No.... ..."

"... In PAGE 3: ... In silico, the 21 exon MCEF gene on chromosome 5 was found to span 88,283 bp, to be transcribed into a 9,580 b mRNA and to code for a 1,163 aa protein, with seven putative NLS, distributed over 4 regions, designated 1/2, 3/4, 5, 6/7 (Fig. 2 and Table1 ). The putative NLS are outside the NHD, ALF, pSer, CHD and a putative transcription activation domain (TAD) regions [3, 24].... ..."

"... In PAGE 4: ...soelectric point are predicted to be 25.8kDa and 8.42, respectively, using Vector NTI software. It is composed of four exons and three introns ( Table1 ), the open reading frame is from 209 nucleotides to 883 nucleotides, the ATG start codon (nucleotides209-211) is preceded by an in-frame stop codon ... ..."

"... In PAGE 2: ...467 In this analysis, we identified such gene structures as shown in Table1 and several unknown structures. As a whole, human genome has much more domain-duplicated genes or protein-coding gene clusters than Drosophila, compared with their gene number ratio.... ..."

"... In PAGE 8: ...able 3-4: Evaluation of promoter prediction systems on the Adh region. .................3-46 Table5 -1: GASP Evaluation of gene finding systems.... In PAGE 8: ...able 5-1: GASP Evaluation of gene finding systems. ...............................................5-83 Table5 -2: Predicted novel genes by Genie.... In PAGE 8: ...able 5-2: Predicted novel genes by Genie. ................................................................5-93 Table5 -3: Genes missed by Genie.... In PAGE 8: ...able 5-3: Genes missed by Genie. .............................................................................5-95 Table5 -4: Joined genes by Genie.... In PAGE 8: ...able 5-4: Joined genes by Genie................................................................................5-96 Table5 -5: Split genes by Genie.... In PAGE 8: ...able 5-5: Split genes by Genie...................................................................................5-97 Table5 -6: Missed long intron(s) by Genie.... In PAGE 8: ...able 5-6: Missed long intron(s) by Genie. .................................................................5-98 Table5 -7: Transposable elements.... In PAGE 8: ...able 5-7: Transposable elements. ..............................................................................5-99 Table5 -8: Alternative splicing forms predicted by Genie.... In PAGE 8: ...able 5-8: Alternative splicing forms predicted by Genie.........................................5-100 Table5 -9: Possible incorrect annotations from std3.... In PAGE 8: ...able 5-9: Possible incorrect annotations from std3..............................................5-101 Table5 -10: GASP evaluation of promoter prediction programs.... In PAGE 8: ...able 5-10: GASP evaluation of promoter prediction programs...............................5-103 Table5 -11: Erroneous EST UTR predictions by Genie.... In PAGE 83: ... Table5 -1: GASP Evaluation of gene finding systems (taken from (Reese et al., 2000)).... In PAGE 83: ...08 In the Adh region, Genie, GenieEST and GenieESTHOM predicted a total of 241, 246 and 258 genes, respectively. In general all three programs scored well in the gene finding category ( Table5 -1, taken from Reese et al. (2000)).... In PAGE 83: ...2.1 Base level The statistical Genie program achieves 96% sensitivity ( Table5 -1). The extra information from ESTs and homology improves the sensitivity of the statistical Genie ... In PAGE 92: ... 5.4 Additional selected observations of the Genie annotation In Table5 -2 twenty-six potential novel genes are listed that are predicted by at least one of the Genie programs and in addition have evidence through an overlap from at least one other gene finding or homology program. The number seems to be very high (over 10.... In PAGE 93: ... Table5 -2: Predicted novel genes by Genie. Twenty-six Genie gene predictions that have no overlaps to any gene structure in std3 are listed.... In PAGE 94: ... Table5 -3 lists the nineteen genes from the reference std3 set for which no overlap of a Genie prediction exists. Thus, less than 10% of the annotated genes in std3 are missed by Genie.... In PAGE 95: ... Table5 -3: Genes missed by Genie. The gene names from Ashburner et al.... In PAGE 96: ...One of the biggest problems with the Genie programs in annotating Adh was joined and split genes. Table5 -4 and Table 5-5 show that Genie is parameterized to favor splitting genes versus joining genes. Only nine Genie annotations span two or more std3 genes (Joined genes) while nineteen std3 genes are split into separate Genie predicted genes (Split genes).... In PAGE 96: ... Another behavior related to the same problem of the length distributions of introns is the general tendency of Genie to introduce erroneous exons within otherwise long introns. Table5 -6 lists eleven typical examples. Table 5-4: Joined genes by Genie.... In PAGE 96: ... Table 5-6 lists eleven typical examples. Table5 -4: Joined genes by Genie. All predictions in which Genie joins one or more genes from std3 are listed.... In PAGE 97: ... Table5 -5: Split genes by Genie. All std3 gene annotations that are split into two or more genes by all three Genie programs.... In PAGE 98: ... Table5 -6: Missed long intron(s) by Genie. Genes that have long introns that are missed by any Genie program are listed and it is indicated which program misses them.... In PAGE 98: ...ransposable elements in the Adh region. Ashburner et al. found seventeen transposable elements, which consist of repetitive elements but also include protein coding like regions, including long open reading frames, predominantly for the transposase and the reverse transcriptase proteins. As expected, Genie cannot distinguish these transposon genes from protein coding genes and therefore predicts thirteen of the existing seventeen as protein-coding genes (see Table5 -7 for a list of predicted transposons). In particular, GenieESTHOM predicts many of the transposable elements to be coding genes because transposable elements contain protein sequences that result in strong protein alignments.... In PAGE 98: ... While the statistical Genie version only overlaps three of the seventeen transposable elements, GenieESTHOM predictions overlap thirteen. These transposon hits contribute to an increased false positive rate, worse wrong exon and wrong gene scores, and lower overall specificity in Table5... In PAGE 99: ... Table5 -7: Transposable elements. Transposable elements incorrectly labeled as real gene by Genie.... In PAGE 99: ... (1999) paper. Str and Begin End Genie Genie EST Genie EST HOM Other gene finder hits Homo logy hits Transposon name F 55,422 58,941 - - X 4 2 Fw R 93,549 94,119 X X X 3 1 G R 255,612 256,662 - - X 1 1 Doc R 959,378 962,797 - - X 2 1 Doc R 1,136,806 1,145,466 - X X 5 0 Roo R 1,293,597 1,298,741 - X X 5 1 Copia F 1,474,114 1,481,634 X (2 genes) - - 3 2 Yoyo F 1,935,760 1,943,170 X X X 3 2 Blood F 2,076,116 2,083,110 - - X 3 2 297 F 2,174,330 2,176,188 - - X 1 1 Copia-like F 2,177,045 2,178,655 - - X 3 2 Copia-like F 2,590,477 2,595,625 - - X 5 2 Copia F 2,603,050 2,610,046 - - X 2 1 297 In Table5 -8 five gene annotations are reported based on Genie predictions that strongly indicate either a different gene structure than reported in std3 or a potentially new alternative splicing form for the listed genes. The underlying evidence, besides the Genie predictions, comes from other gene finding predictions as well as from EST ... In PAGE 100: ...1 Wrong first exon and EST intron in 2nd exon Through evidence from high scoring Genie predictions, EST alignments and the other GASP annotation teams, eight gene entries in the std3 reference set seem to be very suspicious. In Table5 -9, predictions from other programs are only listed if they support the suggested corrected gene structure annotated by Genie. Careful cDNA alignment and additional full-length cDNA sequencing should shed light on these cases ... In PAGE 101: ... Table5 -9: Possible incorrect annotations from std3. Genes from the std3 annotations are listed, for which multiple evidence from Genie and other programs exists, implying incorrect annotations.... In PAGE 101: ...6% for GeniePROM and 32.9% for GenieESTPROM) is in the same order as other promoter predictions from GASP, but indicates that promoter recognition is very difficult due to the complex initiation process (see Table5... In PAGE 102: ... Because of the integration into a gene finding system the numbers should be compared with the similar MAGPIE system and it can be seen that while GenieESTPROM misses two more promoters (31 versus 33) its false positive rate is lower (1/16,786 versus 1/14,760). In addition, in Table5 -10 the prediction statistics for the pure NNPP program are added without the integration in Genie with thresholds of 0.... In PAGE 103: ... Table5 -10: GASP evaluation of promoter prediction programs. We show the number and percentage of identified transcription start sites in comparison to the false positive rate which is given for two different sets of regions: (a) the quot;likely quot; region for a transcription start site plus the downstream region belonging to the same annotation; (b) the same region plus half the distance to the neighboring genes upstream and downstream (taken from the std3 annotation).... In PAGE 103: ... This prescreening masked out these transposon regions and eliminated them from being predicted by Genie. Another structural mistake in the EST based Genie gene models, GenieEST and GenieESTHOM, resulted in erroneous predictions when EST evidence identified introns between non-coding exons ( Table5... In PAGE 104: ... For the Genie application to the entire genome the underlying GHMM gene model was changed by adding the notion of an intron in an UTR region. Table5 -11: Erroneous EST UTR predictions by Genie. Coding gene predictions by GenieEST and GenieESTHOM that are either complete over-predictions or partially wrong by extending the coding regions into the 5 apos;/3 apos; UTR due to a wrong underlying gene model structure for non-coding ESTs (see text for details).... ..."

"... In PAGE 5: ... 3 (a)). On the other hand, genes listed in Table2 (by ISA) did not show any outlier nodes (see Fig. 3 (b)).... ..."